Helicobacter pylori is now recognized as an important pathogen of humans in that the chronic gastritis it causes is a risk factor for the development of peptic ulcer disease. There is also increasing evidence that persistent infection with Helicobacter pylori is a risk factor for the development of gastric adenocarcinoma (1,2), especially of the distal stomach (3-5). The evidence comes mainly from epidemiologic investigations (6-9), including nested case-control studies (4,5,10), and molecular and pathological studies support its biological plausibility (11). However, although essentially all infected persons develop gastritis, clinical consequences of H. pylori infection are recognized in only a minority of persons. Also, while H. pylori infection is highly prevalent in patients with gastric cancer, most H. pylori-infected persons never develop these neoplasms (12). Thus, it is important to identify other factors that more precisely determine risk among H. pylori-infected persons.
H. pylori strains are highly diverse (13-15) at a genetic level, but most phenotypic characteristics are well-conserved. Furthermore, individuals can be infected with more than one strain (4,15). Thus, it is important to isolate particular characteristics of H. pylori strains that might affect risk of gastric cancer development.
Two exceptions to the phenotypic homogeneity are currently recognized. First, about 50%-60% of H. pylori strains produce a vacuolating cytotoxin in vitro (20,39), and toxin production is associated with peptic ulceration (40). Second, there is heterogeneity in whether an antigenic protein migrating at approximately 120-128 kilodalton (kDa) on reducing sodium dodecyl sulfate-polyacrylamide gel electrophoresis [SDS-PAGE] is produced (20). More recent measurements of this protein (referred to interchangably as TagA or CagA) have yielded molecular weights as high as 140 kDa. Although toxic activity is mediated by an 87 kDa protein (23,41), toxin production itself is associated with the presence of the antigenic 120-128 kDa protein (20). Previous studies have found that about 60-80% of H. pylori isolates express the 120-128 kDa protein (20,42). Notably, presence of antibodies to the 120-128 kDa protein in either serum or mucosal secretions is associated with the presence of peptic ulceration (18,20).
Until now, little was known about the association between toxin production, ulcers or gastric carcinoma and the 120-128 kDa antigen. This is due to the previous inability to further characterize the 120-128 kDa antigen after its initial visualization.
In previous studies, the 120-128 kDa antigen was visualized by Western blotting, but virtually no other characterization was performed (20). In contrast to the ease with-which this antigen has been visualized by Western blotting, the 120-128 kDa band has not been easily visualized by other methods such as silver staining (FIG. 2 in Cover et al., 1990 (20)). The explanation for this phenomenon is that this antigen is present only in minute quantities, relative to other H. pylori proteins. Recently, Gerstenecker et al. (43) have reported the isolation of an approximately 120 kDa protein from H. pylori which reacts with positive human control serum. However, virtually no characterization (such as N-terminal sequencing) of this antigen has been performed.
Despite the difficulty of purification, the present invention provides the cloning and sequence of the gene and deduced amino acid sequence encoding the 120-128 kDa protein. This data was obtained using alternate methodology that did not require purification of the 120-128 kDa antigen. The invention also provides diagnostic, therapeutic, and prophylactic compositions and methods.
Immunoblot studies suggest that persons infected with tagA.sup.+ strains have higher degrees of gastric inflammation and epithelial cell damage than do persons from whom tagA.sup.- strains have been isolated (18). Persons infected with tagA.sup.+ H. pylori strains have enhanced expression of IL-1.alpha., IL-l.beta., and IL-8 in gastric biopsies, compared to uninfected persons, and patients infected with tagA.sup.- strains (19). Both intensity of inflammation and epithelial damage may be involved in the pathogenesis of gastric cancer (1). Thus, there is a need to examine the importance of tagA expressing H. pylori strains in this context.
The invention meets these needs by providing a new serologic assay based on a recombinant fragment of tagA, and a method of determining predisposition to gastric cancer.